A distortion correction method and device, electronic equipment and storage medium

By obtaining the actual coordinates of the calibration point and the projection imaging model of the large field-of-view optical camera, a hybrid effect model is constructed for distortion correction, which solves the complex distortion problem of the large field-of-view optical camera and achieves high-precision correction and robust adaptation across the entire field of view.

CN122115282APending Publication Date: 2026-05-29NO 63921 UNIT OF PLA
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
NO 63921 UNIT OF PLA
Filing Date
2026-01-26
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Due to the large field of view, the inconsistent magnification between the edge and center rays, and the assembly factors of optical components, the geometric distortion of large field-of-view optical cameras is difficult to be systematically described by traditional optical distortion models.

Method used

By obtaining the actual coordinates of the calibration points and combining them with the projection imaging model, the principal point coordinates and focal length of the target camera are determined. A mixed-effects model is constructed, including a fixed-effects model and a random-effects model. Distortion correction is performed using a region division and dynamic compensation strategy.

Benefits of technology

It achieves high-precision correction of distortion across the entire field of view, improves adaptability and robustness to the distortion characteristics of different cameras, and is suitable for cameras with large field of view where distortion differences are significant.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122115282A_ABST
    Figure CN122115282A_ABST
Patent Text Reader

Abstract

The application provides a distortion correction method and device, electronic equipment and storage medium, wherein the method comprises: determining actual coordinates corresponding to a plurality of calibration points according to target images obtained by photographing the plurality of calibration points by a target camera; determining principal point coordinates and focal length of the target camera according to the actual coordinates corresponding to the plurality of calibration points; constructing a mixed effect model of the target camera based on the actual coordinates corresponding to the plurality of calibration points, the principal point coordinates and the focal length of the target camera, wherein the mixed effect model comprises a fixed effect model and a random effect model; and performing distortion correction on an imaging area of the target camera through the mixed effect model. The mixed effect model constructed by the application can capture global trends of distortion and correct local residual distortion, thereby achieving high-precision correction of full-field distortion, improving adaptability and robustness of distortion characteristics of different cameras, and being able to adapt to significant differences in regional distortion of large-field cameras.
Need to check novelty before this filing date? Find Prior Art